Similar immunoprecipitation (IP)-based methodologies have been developed for
oxidized forms of 5mC (including 5hmC, 5fC, and 5caC), as the corresponding
antibodies against them are now readily available [83, 84]. This approach has been
used in multiple studies to identify the regions of 5hmC enrichment such as
CpG-rich transcription start sites [83] and 5fC and 5caC enrichment at pericentric
heterochromatin as well as at distal regulatory elements [84].
Apart from direct antibody enrichment techniques, a number of chemically assisted
enrichment techniques have been developed for mapping 5hmC and 5fC. In the anticytosine-5-methylenesulfonate (anti-CMS Seq) technique, 5hmC is converted to cytosine-5-methylenesulfonate (CMS) by bisulfite treatment. CMS-containing DNA fragments are enriched via CMS-targeted antibody and then sequenced [85]. This
methodology has a lower background and lower reliance on 5hmC density in the
fragment as compared to direct antibody pulldown of 5hmC-containing fragments
[85, 86].
In GLIB-Seq (glucosylation, periodate oxidation, biotinylation sequencing),
5hmC is glucosylated in the presence of β-glucosyltransferase, and the resulting
glucosyl derivatives are oxidized with sodium periodate to yield two aldehyde
groups [85]. The latter are derivatized with biotin-containing aldehyde reactive
probe (ARP), and the derivatized fragments are enriched via streptavidin beads
and sequenced, allowing for mapping of 5hmC marks along the genome
[85]. Another chemically assisted enrichment method for 5hmC is hMeSeal-Seq
(hydroxymethyl selective chemical labeling) [87]. In this method, an engineered
glucose moiety containing an azide group is enzymatically added to the hydroxyl
group of 5hmC by T4 bacteriophage β-glucosyltransferase. The azide group can then
be chemically modified with biotin for detection, affinity enrichment, and sequencing of 5hmC-containing DNA fragments [88].
Similar mapping methodologies have been developed for 5fC [89, 90]. Since 5fC
contains an inherently reactive aldehyde group, it can be directly modified with
biotin-containing aldehyde reactive probe (ARP), followed by enrichment of
5fC-containing DNA fragments using streptavidin beads [89]. Another method for
mapping 5fC involves enzymatically blocking 5hmC with unmodified UDP-Glc
[90]. This is followed by 5fC reduction to 5hmC with NaBH 4 and hMeSeal-Seq to
tag, enrich, and map the 5fC fragments as newly formed 5hmC sites [90].
The main limitation of all enrichment-based methods is their low resolution
(100–300 bp) [91]. This can be a major problem, especially for studies investigating
multiple epigenetic marks. Another limitation is that these methods are unable to
quantify absolute methylation levels because each enriched fragment may have
variable numbers of epigenetic marks. Additional common concerns include antibody cross-reactivity, low capture efficiency, and the limitations of downstream
DNA detection methods (microarrays and next-generation sequencing) used to
determine the enrichment. However, these techniques are still widely used as they
provide a rapid, comparatively less laborious and inexpensive way to study genomewide methylation.
504
C. L. Seiler et al.
oxidized forms of 5mC (including 5hmC, 5fC, and 5caC), as the corresponding
antibodies against them are now readily available [83, 84]. This approach has been
used in multiple studies to identify the regions of 5hmC enrichment such as
CpG-rich transcription start sites [83] and 5fC and 5caC enrichment at pericentric
heterochromatin as well as at distal regulatory elements [84].
Apart from direct antibody enrichment techniques, a number of chemically assisted
enrichment techniques have been developed for mapping 5hmC and 5fC. In the anticytosine-5-methylenesulfonate (anti-CMS Seq) technique, 5hmC is converted to cytosine-5-methylenesulfonate (CMS) by bisulfite treatment. CMS-containing DNA fragments are enriched via CMS-targeted antibody and then sequenced [85]. This
methodology has a lower background and lower reliance on 5hmC density in the
fragment as compared to direct antibody pulldown of 5hmC-containing fragments
[85, 86].
In GLIB-Seq (glucosylation, periodate oxidation, biotinylation sequencing),
5hmC is glucosylated in the presence of β-glucosyltransferase, and the resulting
glucosyl derivatives are oxidized with sodium periodate to yield two aldehyde
groups [85]. The latter are derivatized with biotin-containing aldehyde reactive
probe (ARP), and the derivatized fragments are enriched via streptavidin beads
and sequenced, allowing for mapping of 5hmC marks along the genome
[85]. Another chemically assisted enrichment method for 5hmC is hMeSeal-Seq
(hydroxymethyl selective chemical labeling) [87]. In this method, an engineered
glucose moiety containing an azide group is enzymatically added to the hydroxyl
group of 5hmC by T4 bacteriophage β-glucosyltransferase. The azide group can then
be chemically modified with biotin for detection, affinity enrichment, and sequencing of 5hmC-containing DNA fragments [88].
Similar mapping methodologies have been developed for 5fC [89, 90]. Since 5fC
contains an inherently reactive aldehyde group, it can be directly modified with
biotin-containing aldehyde reactive probe (ARP), followed by enrichment of
5fC-containing DNA fragments using streptavidin beads [89]. Another method for
mapping 5fC involves enzymatically blocking 5hmC with unmodified UDP-Glc
[90]. This is followed by 5fC reduction to 5hmC with NaBH 4 and hMeSeal-Seq to
tag, enrich, and map the 5fC fragments as newly formed 5hmC sites [90].
The main limitation of all enrichment-based methods is their low resolution
(100–300 bp) [91]. This can be a major problem, especially for studies investigating
multiple epigenetic marks. Another limitation is that these methods are unable to
quantify absolute methylation levels because each enriched fragment may have
variable numbers of epigenetic marks. Additional common concerns include antibody cross-reactivity, low capture efficiency, and the limitations of downstream
DNA detection methods (microarrays and next-generation sequencing) used to
determine the enrichment. However, these techniques are still widely used as they
provide a rapid, comparatively less laborious and inexpensive way to study genomewide methylation.
504
C. L. Seiler et al.